1. A study of mosquito breeding in plant axils was carried out in Bwamba County, an area with much heavy forest in the extreme west of Uganda. Attention was directed mainly to Aëdes (Stegomyia) simpsoni, a known vector of human yellow fever in Bwamba.

2. The various plants studied are described.

3. The results of preliminary surveys are discussed. These confirmed that, as far as mosquito breeding is concerned, the most important plants in Bwamba are the lowland colocasia (Xanthosoma sagittifolium), the “gonja” variety of cultivated banana and the pineapple (Ananas comosus). It was shown that A. simpsoni is the commonest axil-dwelling larva in the inhabited lowland areas, and that it occurs in widely different types of lowland country. Records are given of the occurrence of larvae of Aëdes (Stegomyia) aegypti in plant axils.

4. A survey of 12 different kinds of plants is described, and it is shown that the larval populations depend to a considerable extent on the amounts of water retained by the axils of the various plants concerned. In Bwamba, medium-sized axils with a capacity of 4 to 8 cc. appear to be the most favourable. Certain plants which yielded large numbers of larvae—such as the wild banana (Musa sp.)—were not found to harbour A. simpsoni. Apparently this is due to the fact that they are confined to forest and mountain areas, where A. simpsoni is scarce or absent.

5. A monthly survey of Xanthosoma axils, carried out over a period of 12 months, showed that A. simpsoni larvae were continually present and that the fluctuations in larval population were dependent on growth and harvesting of the plants rather than on rainfall.

6. Scrapings from dry Xanthosoma axils, taken during a short period of drought, were placed in bowls of water in mosquito-proof cages. Larvae of A. simpsoni and Harpagomyia taeniarostris hatched from this material in 2 to 3 days.

7. Temperature measurements in a Xanthosoma axil showed that the water therein was cooler than the surrounding air and also that the axil temperature was remarkably equable.

8. Systematic collections of pupae from Xanthosoma axils showed that the mosquito output (which cannot be gauged directly from the population of larvae) is high and constant. It is estimated that a plot of 500 plants of this species may, under favourable conditions, produce something like 5,000 A. simpsoni (2,000 females) per month.

The influence of predatory larvae (Eretmapodites spp.) on the populations of other species is thought to be considerable.

9. The fauna of plant axils, apart from mosquito larvae, is described briefly.

The importance of finding the effect of defoliation on the reproductive capacity of Cordia macrostachya is explained in view of the proposed introduction of two species of leaf-eating beetles from Trinidad into Mauritius for the control of this weed.

An account is given of experiments in which the production of inflorescences and fruits was compared in defoliated and control plants.

The results show that defoliation has a significant effect in reducing the reproductive capacity of the plant.

Cordia macrostachya (Jacq.) R. & S. (Boraginaceae) is a serious pest in Mauritius, to which island it was accidentally introduced from the West Indies. During investigations of the possibility of the biological control of this weed, Kirby (1944 unpublished report, “Insects of Cordia in Trinidad”) and Donald (1945 unpublished dissertation) worked out the biology of the Galerucid, Schematiza cordiae, Barber (J. Wash. Acad. Sci. 37, p. 243, 1947), a defoliating species that gave promise of meriting consideration for introduction into Mauritius.

In December 1942 and January 1943, the late Captain T. P. Ward, R.A.M.C., carried out a malaria survey of the island of Socotra accompanied by Private G. F. W. Hart, R.A.M.C., who collected the mosquito material. The following notes have been abstracted from Captain Ward's official report.

It is now more than ten years since the publication of Harris' list of insects injurious to native food crops in Tanganyika (Harris 1937). It may be well to recall the view expressed in his introduction “… these crops … have achieved, within the limits of their present distribution, an equilibrium with their insect pests. Where there is no reasonable return for his labours, the native has ceased to grow a particular crop. This equilibrium is unstable, since pests increase in numbers at intervals owing to a variety of causes …”.

(1) The Kabaw and Kale Valleys, separated by a low watershed and both draining through lateral gorges, have a combined length of some 250 miles and a width of between 4–10 miles. They lie at an altitude of approximately 500 feet above sea level.

(2) There is heavy rainfall during the months of June to September inclusive and little or no rain in the months November to April inclusive. The Kabaw Valley, the more northerly of the two, is forested and there are only small isolated clearings, but a big area in the centre of the Kale Valley has been cleared. Rice is the main crop grown.

(3) There appears to be no pre-war information about mosquitos and malaria. Information during the war was gathered by several different types of Unit engaged on different types of work but there is sufficient to give a fairly complete picture.

(4) The valleys are highly malarious throughout their length and breadth; only one village out of twenty-three examined had a spleen rate under 60 per cent.

(5) Transmission of malaria takes place throughout the year and is intense during the months May to December inclusive.

(6) 17 species of Anopheles are recorded. Common species during the rains are A. minimus, A. leucosphyrus, A. philippinensis, A. jeyporiensis, A. annularis, A. hyrcanus and A. vagus. A. minimus is only locally abundant at this time; A. leucosphyrus is widespread but confined to the forest.

(7) When the rain ceases, species such as leucosphyrus and vagus, which breed in pools and similar habitats, become reduced in numbers. Species such as philippinensis, jeyporiensis, annularis, hyrcanus and barbirostris, which breed in larger bodies of stagnant water such as rice fields, also become reduced in numbers during the dry season, though more slowly, because their breeding places disappear less rapidly. As the streams and rivers settle down, after the rains have stopped, conditions for A. minimus and A. maculatus become good and these are the only two common and widely distributed species in the dry season.

(8) During the dry season, sporozoite rates of A. minimus were 1/159=0·63 per cent, in October, 19/519=3·66 per cent, in December and 10/514=1·94 per cent. in May and it is an important vector and the only vector. A. maculatus does not apparently attack man and plays no part in malaria transmission. In September the sporozoite rate of A. minimus was only 1/332=0·31 per cent, and the important vector of the period of the rains appeared to be A. leucosphyrus with a sporozoite rate of 2/134=1·49 per cent.

(9) The status of A. leucosphyrus as a vector in the Indo-Burmese border region is discussed. There is good evidence that, under wartime conditions with troops operating in the forest, it is an important vector during the rains. Its significance in peacetime is less easy to assess; as only a few people live in the forest or at the forest's edge, its importance is probably not great. On the other hand it does not rest indoors, and, as in the past reliance has been placed on day catches, it may have been overlooked. There is, however, evidence that military operations in the forest produce conditions which lead to an increase in the numbers of this species.

(10) Adult collections were nearly all made at night in camps or villages. The species found in large numbers near man were A. minimus, A. leucosphyrus, A. philippinensis, A. jeyporiensis, A. annularis, and A. hyrcanus. A. minimus and A. leucosphyrus, the two definitely incriminated vectors, were the most successful at obtaining blood when the host, man, was protected by a mosquito net. The other four species, however, were sometimes filled with blood, undoubtedly of human origin and all may play some part in malaria transmission under war conditions. The chances of a specimen becoming infected were reduced under the conditions existing when the dissections were made and so moderately long series of negative dissections (Table XI) cannot be taken as conclusive evidence that any one species played no part in malaria transmission. In peacetime these four species may feed mainly on cattle, though a description is given of a night catch in a village where A. philippinensis was caught attacking man although cattle were available.

(11) In September very few of any of the eight common species were resting by day in the place where they fed at night and it is presumed that they spent the day in the forest. Unfortunately very few inhabited houses were visited and almost the only available catching places were Army tents and shelters. The results suggest, however, that all the species tend to rest in the forest rather than indoors. In December in the Kale Valley A. minimus rested in houses by day; in May in the Kabaw Valley it did not. It is believed that where, as in the Kabaw Valley, forest comes up to the edge of the village, A. minimus seeks shelter for the day in the forest, but that where, as in the Kale Valley, the village stands in a clearing, it rests by day in the houses near where it has fed. The facts observed, however, do not preclude the possibility of a seasonal change in resting habit. It is clear that A. minimus is not invariably a house rester and the importance of catching by night to gain a true picture of the mosquito population is stressed.

(12) A. hyrcanus was usually the first species to appear after dark, followed by A. philippinensis, A. leucosphyrus and A. minimus in that order, the last named being usually most abundant in the second half of the night. Time of greatest activity was, however, variable in all species. There were only a few isolated instances of Anophelines biting in the forest by day.

(13) A. hyrcanus was attracted to light.

(14) Larvae identified as A. pallidus with the aid of the standard Indian keys invariably proved to be A. philippinensis on emergence.

(15) The variation found in A. minimus was considerable and is briefly discussed. The only constant difference between A. minimus and A. aconitus was found to be the flavescence on the proboscis.

DDT and “Gammexane” have many advantages over arsenic as acaricidal agents.

The sites of attachment of the common tick species on bovines are recorded and the killing properties of various dilutions of DDT and “Gammexane” are assessed by observations on R. appendiculatus attached on the ears and A. variegatum attached on the udder and flanks of cattle.

The preparations were usually applied as sprays by means of a flit-gun, an Aerograph spray gun, and Kent sprayer or “Solo” hand pump. A “swirling-jet” nozzle was found most suitable. A possible combined spraying-dipping method using a shallow bath is also described.

In preliminary experiments, a spray containing 2·5 per cent. DDT and a preparation containing 1:10 of the “Gammexane” dispersible powder P.530 killed all ticks attached at time of spraying but the residual effects were lost about the 4th day after spraying. Field experiments gave similar results, but 1:20 of the P.530 powder (0·325 per cent. active gamma) proved equal in killing value to the 1:10 preparation.

The addition of coumarone resin did not increase the residual effects of a 2·5 per cent. DDT emulsified solution.

When DDT and “Gammexane” were first used in Uganda for work on tick control, the absence of a rapid and simple chemical test made it necessary to evolve a biological method which would assess the killing power of the preparations in use. Such a method, if it is to be employed to assess the results of field work, should satisfy the following conditions;

(a) ticks should receive an initial wetting with the preparation comparable to that received when cattle are dipped or sprayed, and

(b) remain on a moist surface for a few hours, and then for a longer period in contact with the dried deposit, as in the case of the animal after it has been dipped or sprayed.

The parasitic larvae of a number of Trombiculid mites are of economic importance as they include the vectors of scrub typhus, and scrub-itch mites of considerable nuisance value. Methods of trapping these mites and various field and laboratory observations on them are described. The methods were evolved and observations made during the course of work, sponsored by the Medical Research Council, in Imphal, Assam and Compton, Berkshire.

Live unfed larvae in the field may be concentrated and trapped by use of a low cover with a central light-admitting aperture and collecting apparatus. This method proved very efficient in trapping harvest mites (Trombicula autumnalis) in England. The rise in temperature within the trap can be made advantageous.

Guineapigs were chosen as bait animals for the collection of T. autumnalis at Compton, most of the larvae attaching near the anus and genitalia. The larvae fed for between 30 and 60 hours.

Engorged larvae may easily be collected as they leave their host (or dead animals and débris) by the funicular laboratory light-trap described.

Observations are made on the collection of adult mites, and to a lesser extent nymphs, from soil flotation in water. The vertical migration of adults in relation to soil moisture and temperature is discussed.

Observations on the appearance of larvae of T. autumnalis on a chalk-down at Compton showed a sharp seasonal peak in September, 1946, and some evidence is adduced that this was related to a peak in oviposition.

Observations are made on attempts to rear T. deliensis, the scrub typhus vector in Assam, and notes are added on the behaviour of larval and nymphal mites.

The genus Stethorus Weise is of special interest on account of its cosmopolitan distribution and the habit of its species of feeding mainly on mites, some of which are pests of crops, including fruit trees. Its relationship with other members of the tribe Scymnini is also of importance, as it was separated by Dobrzhansky in 1924 into a distinct tribe called Stethorini, which, however, was regarded by Korschefsky (1931) as synonymous with Scymnini. The material on which the present study is based came mainly from the Old World. Much of it had been submitted to the Commonwealth Institute of Entomology for identification, and I am grateful to Dr. W. J. Hall and Dr. T. H. C. Taylor, of this Institute, for giving me facilities and help. The remainder of the material studied was in the collection of the British Museum. I am indebted to Mr. N. D. Riley for procuring, on my behalf, certain type specimens from abroad for examination.